Xfce Forum

Sub domains
 

You are not logged in.

#26 2020-01-27 22:13:08

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

Aravisian wrote:

When I first started using XFCE4 desktop, I noticed that problem right away. So, on my themes, I went into the XFWM4 folder and adjusted those images accordingly, giving them more space (Usually 9px) while leaving the Visible portion at 1px.

Do you have an example? I had a look at AzenisXFM-Purple and it seems to have the 5px resize area suggested by the graphics files, which is a couple of pixels narrower than the border in the Xfwm4 theme I tend to use, which is 4deb: https://www.xfce-look.org/content/show. … ent=166997

The compositor approach works without additional user/dev input if available, as a write-once-use-almost-everywhere fix. That sort of consistency is useful for less technical users.

Aravisian wrote:

You have a knack for concisely making a strong point. I thoroughly enjoyed reading the Blog link posted earlier.

Thanks, it's probably the day job overlapping. I usually write to organise thoughts, but if anyone else gets any use/benefit that's always nice to know.

Offline

#27 2020-01-27 22:34:19

File Manager
Member
Registered: 2019-11-01
Posts: 62

Re: Do you want to use CSD? No please!!

denyer wrote:

It's the Gnome way. I only found Xfce about three years ago, when Windows became too awkward to be using outside of work, and knowing that it's survived since 1996 without major changes was part of the appeal -- it's a shame to see existing functionality cut and vandalised.

denyer wrote:

In Xfce, rather than make it so so that users have the option of using a compositor and having a workable resize area independent of theme, nobody gets one unless they pick a suitable theme. Or unless they opt to lose other window manager functionality and use CSD apps. Or replace the window manager.

This is particularly a problem with small development projects; decisions can end up being taken for ideological or vanity reasons rather than with input from across the user base and with regard to the impact on accessibility.

I wasn't just being bullish by describing the approach as vandalism. It's been deliberate and avoidable.

The lack of honesty from Xfce developers is troubling. They are smart people and realize that CSDs are a regression from a user perspective. Also, what's up with those symbolic icons breaking all old icon themes? It's GTK3 madness breaking everything and this will of course continue with GTK4 and Wayland. I don't see Windows and macOS reinventing the wheel and breaking everything in the process.

GTK3/4 and Wayland may force me back to proprietary OSes and I think that was the idea behind these corporate projects. A desktop OS is not a smartphone OS even if Gnome thinks it should be.

Offline

#28 2020-01-27 22:56:46

oldhwman
Member
Registered: 2019-10-07
Posts: 6

Re: Do you want to use CSD? No please!!

Just wanted to throw a quick thought into the wind here. I'm not a big fan on header bars (for reasons I've mentioned in this topic) but it's worth pointing out that, with Gnome being such a big player, and GTK being developed, essentially, by the Gnome project and for the Gnome project, a lot of GTK applications are using CSD and header bars, for better or for worse. If you're using GTK applications, they're pretty hard to avoid by now. There aren't many developers writing applications for Linux, and many of those who do write applications that target Gnome. Mate and Cinnamon aren't doing great in terms of developers' time, either, so it's probably not a good idea to bet on "classic" applications (e.g. Atril) being around forever. Adopting these things in Xfce means more applications will fit in. Maybe it's not the best way to do that but who knows. Linux doesn't exactly have a huge pool of applications that you can use. Being able to use only a few applications from an already small set of programs is only making things worse. Making more applications fit in means a wider pool of applications for Xfce users, even if the approach is kindda backwards.

Realistically, practical experience from other platforms shows that most users would rather put up with bad design choices but have more applications -- and thus, more functionality and/or more alternatives -- than fewer applications that look just right. I mean, I don't like headerbars and hamburger menus, but if the only application that does what I need has headerbars and hamburger menus, I'm gonna put up with them...

So yeah, maybe it looks bad, but at the end of the day it's not just about how it looks in screenshots. I can see how this decision might in fact benefit a lot of users.

On a more speculative note: I do think it's a little late to jump on the headerbar bandwagon. The lukewarm reception that "universal" interfaces have met everywhere except in the consumer space has been pretty discouraging. We've been pushing them onto users for like eight years now, touting them as the biggest revolution since sliced bread, but virtually everyone who is not a designer, a tech evangelist, or a consultant has been, at best, indifferent. It looks like the interfaces we've been coming up with for the last eight years or so are great if you want to watch cat gifs and share links to really influenshial, like, food for thought, man, Medium posts, but they kindda suck if you have to get any real work done. Companies with paying customers will put up with a stupid choice only for so long before they get their designers in line and stop shoving something down their customers' throat. And this shapes design choices slowly, but surely. There are already a lot of voices in the designer community that are beginning to question the wisdom of exposing the same interface on phones, tablets and desktops. Adopting headerbars would have been hot in 2013. Today it's more or less the equivalent of adding an animated assistant in a word processor in 2000 -- granted, someone did it a few years ago, and all the cool kids are doing it now, but secretly, even die-hard Clippy fans are beginning to suspect it's really, really stupid, and ten years from now we'll want to kick ourselves for doing it.

However, I also doubt that the way forward is designing the same kind of interfaces we've been designing twenty years ago. This bandwagon may be a bit too 2013, but I suspect the next bandwagon to jump on looks more like this one than like the Amiga. So it may be too late to be a cool move, but still not too late to miss the next train.

Last edited by oldhwman (2020-01-27 23:01:00)

Offline

#29 2020-01-28 01:54:51

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

oldhwman wrote:

If you're using GTK applications, they're pretty hard to avoid by now.

Chrome and other browsers do it and tend to be liked for it, but that's a rare example of utility. Certain headerbar apps reliably crop up in distros other than Gnome, such as Disks where maintainers see value in something simpler than GParted.

Better known FOSS couldn't really care less about headerbars -- VLC, LibreOffice, Audacity, GIMP, FileZilla, etc. Software where most of the users aren't on Linux but on Windows, where the interfaces largely come down to traditional menus or (rarely) Microsoft's ribbons, which haven't gained wide credence either.

oldhwman wrote:

practical experience from other platforms shows that most users would rather put up with bad design choices but have more applications -- and thus, more functionality and/or more alternatives

Also would tend to disagree -- there are projects that are clear leaders and which have risen above others for various needs; the value of developing applications such as Parole is limited when most people will go straight for VLC, most distros bundle LibreOffice, etc. More commodity apps such as PDF readers tend to involve the same/similar fairly feature complete code with minor tweaks, and as long as there's a toolkit they can rework and compile against they aren't going to radically re-engineer their interfaces.

The largest projects are unlikely to let toolkit developers get away with dictating, which helps to preserve choice for smaller projects.

At the moment GTK3 has increased leverage due to the tie-in with Wayland development. GTK4 adoption is likely to be even slower and contingent on it adding back in things people actually want. Ultimately if it goes too far, cleverer folk will get frustrated enough to fork, keep the worthwhile bits, work on the rest and effectively sideline the original project as happened with OpenOffice.

I'm fairly optimistic. There'll be a fair amount of upset and drama, there always is, but I don't think Gnome will win. As you say, universal interfaces haven't exactly sold themselves.

oldhwman wrote:

I also doubt that the way forward is designing the same kind of interfaces we've been designing twenty years ago.

We're still using metaphors and layouts based on paper, mice that don't need batteries, keyboards with key arrangements designed to slow people down, etc. Some things have longevity and tend to resist "solutions" from hip young designers looking for problems to fix. Voice input probably isn't going to take over, either.

Offline

#30 2020-01-28 06:48:50

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

denyer wrote:

Do you have an example? I had a look at AzenisXFM-Purple and it seems to have the 5px resize area suggested by the graphics files, which is a couple of pixels narrower than the border in the Xfwm4 theme I tend to use, which is 4deb: https://www.xfce-look.org/content/show. … ent=166997

I don't off the top of my head, but I can make you one if you like.
I have never had any trouble grabbing the edge even on only 5px. No one has commented on having any trouble, so I didn't bother modifying it.
But I can make a Modified Border (Preferably a 1px wide theme ) and offer it up for a test. Admittedly, I would prefer to not upload it to Pling and cause Confusion about what that is... but I can email you a custom made window border.
For the moment, let me explain how to do it graphically (the easy way without doing any coding). Take a close look at that Azenis Purple window Border that you referenced. This is a good one because it shows what I am describing neatly.
On the right side, you have the Close Button. Then you see some shadow, a line, then a couple of millimeters of border edge with a rounded top, right? Below that, in the theme, you would see the menubar. In the way that window border is structured, that menubar is inset slightly and you see where the right-border meets the Titlebar, that the border is further out than the line in the shadow next to the close button. So that border looks about (a little over) a millimeter wide and the portion at the top looks about two mils wide.
Azenis was created by J.J.Ying for Windows and Window Blinds and part of theming it in Linux (He granted his permission) included adapting the theme. This is a Known Part of Theming. It will not line up the same way- so you must adapt.
Instead of just using the border image as supplied, the trick is to set your canvas to the width you want first. So let's say we want a 10px border, we set that as our canvas. The image is 5px, so we set that on the outside edge of our canvas. Now, to adapt our Windowblinds imagery to XFWM4, we draw the inner portion of our theme background in the remaining space. This is because a solid pixel is needed to grab (Though I have done this with opacity, as well and that gets more complicated - I want THIS answer to be the easiest one to describe).
So the border image, sized, in your XFWM4 folder would actually look like the inner portion of your theme background and then the exterior border line, whether it is 5px as in this example using Azenis or 1px for a different example.
Let's say we have a window border with a 1px black line and the inner theme is a white background. Then the border in the XFWM4 folder would look like a 1px black line with 9 pixels of white space next to it-- Instead of the themer just putting the border as a 1px black line image.
A themer that was used to a different window manager would not think to include the inner portion of the theme background and so- we ended up with many window borders made that did not account for grabbing space.
In spite of how wordy the above was- you can see how it breaks down to something very simple to do- but you can also see how many people wouldn't think of doing it.


This topic is creating a lot of discussion and I am not just enjoying the debate and various points of view but also have learned several new things. I never knew about gtk-nocsd until this topic and that led to a whole new rabbit hole to fall into.  I did not know Gnome was removing Module Support and that is a very big deal.

File Manager wrote:

It's GTK3 madness breaking everything and this will of course continue with GTK4 and Wayland.

It is interesting how Wayland is being pushed so very hard while it just doesn't work at all and has such poor compatibility.

File Manager wrote:

I don't see Windows and macOS reinventing the wheel and breaking everything in the process.

VERY TRUE.
Not to promote Windows (especially with all that has been in the news about it with security issues), but even with many improvements to UI made, the base foundation of it remains the same. Going back well over 10 years. They never broke that.

File Manager wrote:

GTK3/4 and Wayland may force me back to proprietary OSes and I think that was the idea behind these corporate projects. A desktop OS is not a smartphone OS even if Gnome thinks it should be.

Gnome talks about it's 'vision.' Gnomes vision is Complete Control. Gnome is constantly changing things trying to create a Unique Gnome experience and to drive out others.
In the above post, you made some very excellent and strong points. No need to quote all of it just to cheer, however this line:

denyer wrote:

I'm fairly optimistic. There'll be a fair amount of upset and drama, there always is, but I don't think Gnome will win.

I wish I shared your optimism. Sadly, it looks to me like Gnome will win simply because they have seized enough control of the board. Even many tech articles put forth strangled apologetics for Gnome, trying hard to convince us that the medicine is oh-so-delicious.

Last edited by Aravisian (2020-01-28 06:52:08)

Offline

#31 2020-01-28 10:14:27

oldhwman
Member
Registered: 2019-10-07
Posts: 6

Re: Do you want to use CSD? No please!!

denyer wrote:
oldhwman wrote:

If you're using GTK applications, they're pretty hard to avoid by now.

Better known FOSS couldn't really care less about headerbars -- VLC, LibreOffice, Audacity, GIMP, FileZilla, etc. Software where most of the users aren't on Linux but on Windows, where the interfaces largely come down to traditional menus or (rarely) Microsoft's ribbons, which haven't gained wide credence either.

I was specifically referring to GTK applications. VLC, Audacity and Filezilla don't use GTK, LibreOffice does its own convoluted thing, and Gimp's GTK 3 port isn't there yet (plus it's still trying to be cross-platform).

But Gnome has a lot of leverage in this space. New GTK 3 applications are likely to do the whole headerbar thing simply because the project behind the toolkit is promoting them, and it's the environment that most users are likely to run. If you're just learning GTK, you're going to learn it from Gnome's tutorials, and look at source code from the Gnome project. You'll read the Gnome HIG docs while learning to build interfaces with GTK. The first generation of programmers who was exposed to nothing but Windows 8-style interfaces is just coming of age, and they're going to be writing most of the FOSS code that's gonna get written in the next ten years, since they have the most free time. Those of us who grew up on Blitz and Visual Basic aren't gonna be around writing software forever.

oldhwman wrote:

practical experience from other platforms shows that most users would rather put up with bad design choices but have more applications -- and thus, more functionality and/or more alternatives

Also would tend to disagree -- there are projects that are clear leaders and which have risen above others for various needs;

...and it's all fine and dandy until they get improved to death, like Adobe Reader, or Kmail, or Gedit. There's limited value in developing less popular applications until the popular option turns weird. In the last days of Gnome's 2.x, lots of people didn't see the point of XFCE, either.

The largest projects are unlikely to let toolkit developers get away with dictating, which helps to preserve choice for smaller projects.

The largest projects don't have a choice. You don't just port a 400 kloc codebase over from GTK to Qt in a few evenings. Just moving from GTK 2 to GTK 3 is a significant effort, and the two are similar enough that the process isn't radical, just very, very tedious. Even that took years for some projects. Even if your code is very well structured, a la Wireshark, it still takes a lot of time. If toolkit developers make a choice, you get to live with it. Headerbars are a special case because you can always just not add one to your application, but changes to a toolkit's core are things you get to adopt whether you want them or not.

I, too, hope these things die a fiery death but I don't think their fiery death is just around the corner smile.

Offline

#32 2020-01-28 21:12:14

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

Aravisian wrote:

Let's say we have a window border with a 1px black line and the inner theme is a white background. Then the border in the XFWM4 folder would look like a 1px black line with 9 pixels of white space next to it-- Instead of the themer just putting the border as a 1px black line image.

Ah, get you. I got confused by the visible portion part, coming from a background that's more web interfaces (although I used to theme things that were bitmap-based like Winamp). More theme designers could definitely work around the issue with inner margins (padding in stylesheet terminology) rather than relying on margins outside of the window in the area a compositor would typically draw a shadow.

And if people are specifically creating Xfwm4 themes, it's reasonable to expect them to take that approach. With GTK themes the expectation is more that the window manager will supply the resize areas.

Aravisian wrote:

This topic is creating a lot of discussion and I am not just enjoying the debate and various points of view

Definitely agree -- always useful to pause, evidence things and get more information.

Aravisian wrote:

Sadly, it looks to me like Gnome will win simply because they have seized enough control of the board. Even many tech articles put forth strangled apologetics for Gnome, trying hard to convince us that the medicine is oh-so-delicious.

It certainly is noticeable that articles mentioning Gnome in Ubuntu/Fedora tend to assume obligatory defensive caveats, or point out that the first thing they do with a fresh install is add tweaks or other software to get back to a more traditional desktop environment.

Aravisian wrote:

I did not know Gnome was removing Module Support and that is a very big deal.

The gtk3-mushrooms approach seems to be to patch/fork the toolkit directly; i.e. a user wouldn't have gtk3 and gtk3-classic installed simultaneously.

Forking is a big commitment, or even trying to get patches accepted to maintain previous functionality, but the fact that it's possible with FOSS and that a lot of other sufficiently technical people are going to be upset provides some hope on that score.

One of the biggest issues for Linux is that when significant changes in approach are made, there are often protracted periods of upheaval -- by the time I came back to Linux, systemd was fairly established, to pick the usual example, and had been through a lot of the initial breakage of established ways of doing things. Functionality that was missing when it was originally dropped into major distributions by eg the Debian vote has, in at least some cases, since been added. The initial systemd implementation of hibernation wouldn't have worked with my main machine, for instance, it needs 'shutdown' mode.

Overall, that uncertainty and those periods of upheaval or breakage tends to harm the user experience of desktop Linux, and leads to the usual advice for people to follow LTS releases and various snap/flatpak/etc workarounds emerging for acquiring fresher packages (personally I use some appimages and extract-and-run software but otherwise normally prefer six-monthly releases as a compromise between LTS and rolling release; I could probably be at home with an Arch-based distro, but Ubuntu derivatives are convenient).

But because there are so many technically-minded users, there tends to be a recovery -- Gnome 2 returning as MATE, or people finding alternatives or workarounds for GTK issues.

Personally as I've said, I'm not greatly attached to theming and if window management functionality isn't lost by a broken CSD implementation (which could be done better, but just hasn't been) then uses/projects like KDE demonstrate some can be made to better integrate visually. And I think the missing functionality such as variant maximising will make a comeback, and continuing to mostly avoid CSD/headerbar apps for the time being is going to be the best approach. It's a great shame that that means Xfce becomes a dead end for now.

Theming in the traditional sense will likely continue to be harmed by the need/desire for more users to have HiDPI hardware support -- themes that use vector-based or high resolution graphics, or want to target touchscreens, are more complicated than traditional bitmap-based stuff and it becomes an area where interested dabblers can't compete with the usage scope addressed by themes like Breeze or Adwaita.

oldhwman wrote:

I was specifically referring to GTK applications. VLC, Audacity and Filezilla don't use GTK, LibreOffice does its own convoluted thing, and Gimp's GTK 3 port isn't there yet (plus it's still trying to be cross-platform).

The big projects are going to stay and want to stay cross-platform, and if GTK doesn't easily support that people will go with things that do. People won't willingly start projects with GTK, and will join a number of developers I've seen cursing the fact that they picked GTK in the first place -- some projects will drink the kool-aid, some will switch, and some will be abandoned rather than make the time investment.

The situation's most complicated for projects with existing big investments, caught in the upheaval and uncertainty. As you say, it can be a big commitment (or project-killing). At a DE level it's more of a radical choice and you get situations like LXDE and LXQt. Xfce isn't going to go that way, it's GTK (or a drop-in fork) or nothing.

oldhwman wrote:

The first generation of programmers who was exposed to nothing but Windows 8-style interfaces is just coming of age, and they're going to be writing most of the FOSS code that's gonna get written in the next ten years,

I honestly think that the divide between mobile and desktop interfaces is only going to become more entrenched, mobile-everywhere having been found wanting. Microsoft has continued to dial back the Win8 stuff after experiencing a second Vista-type reaction and found ways to support different hardware or different uses of the same hardware differently, and presently shifted to putting its efforts into catching up with Android/iOS on selling user data and trying to force people down subscription routes and/or to web apps. Although there are more computer users globally, a smaller proportion are actually coding and a lot of those that are developing focus on web interfaces.

I'm still optimistic that even if we won't be around forever, we've passed torches to enough kids to demonstrate that some interfaces are more productive for relatively complex productivity apps.

oldhwman wrote:

until they get improved to death, like Adobe Reader, or Kmail, or Gedit

Not a new cycle, and there are still decent alternatives. Gedit in particular is commodity software and for knocking up a short text file is still far better than Notepad, but IDEs and editors are resilient if any good and they have a following. Emacs / Notepad++ / Visual Studio aren't likely to be ruined, and the FOSS options tend to reroute around attempts to. Even with Gedit more people cared enough to salvage Pluma out of it than apparently cared to maintain Gedit.

I'm a bit more worried about Thunderbird, although it seems to be hanging in there for now.

Offline

#33 2020-01-28 22:42:29

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

denyer wrote:

And if people are specifically creating Xfwm4 themes, it's reasonable to expect them to take that approach. With GTK themes the expectation is more that the window manager will supply the resize areas.

You got it, exactly. I was worried I may end up explaining my point of view on this again. And I stress, it is a hypothesis, not a 'fact.'
But it does seem extremely likely to me that the Resize Issue is more a misunderstanding than it is a failure. Yes, the initial theming of XFWM4 was performed without the themer realizing that he or she needed to adapt to the process instead of directly putting in just the original image.
Admittedly, I only did it on those which had very small borders... and even then, I looked at what I have made earlier and realized I missed on the GT3 theme. I will have to go back in and correct that.
So, you have my gratitude for helping me spot that.
EDIT: I am editing this in a bit late but oh well...
I just had one of those things... It's like a headache but it gives you ideas... Oh yeah- A thought.
I often use an eye-candy Cursor. Currently, it is "The Array". https://www.deviantart.com/grynays/art/Array-308857373
The larger Animated Cursor actually makes it a lot easier to grab a resize area than a small dinky not-animated cursor. Because the cursor begins to move and align itself when nearing the grab area and because it is larger, it makes it easier.  Another point is to slow the cursor speed so it doesn't fly across the screen when you twitch your hand on the mouse. Just thought... I'd throw this out there... and go get myself an aspirin.---END EDIT.

denyer wrote:

The gtk3-mushrooms approach seems to be to patch/fork the toolkit directly; i.e. a user wouldn't have gtk3 and gtk3-classic installed simultaneously.

I have had no luck getting that to work. I have tried- I cannot grow shrooms, it seems.
I am using gtk-nocsd for now. I AM big on theming. When you spend enough time at the computer, you want to feel JOY at the eye candy when you have to stare at it instead of looking at a dull, flat, drab, uniform screen. Knowing my screen looks vastly different from most and seeing peoples eyes bug out (Mostly Windows users) when they see it makes it bearable.

denyer wrote:

But because there are so many technically-minded users, there tends to be a recovery -- Gnome 2 returning as MATE, or people finding alternatives or workarounds for GTK issues.

Point:
Mate was a continuation of Gnome 2 using gtk2 for a short time but Mate is now fully GTK3. 100%.
The surface looks like gtk2 but it is a facade, under the hood it is all Gnome 3.
However, the Mate Panel and the Marco WM both are well designed contributions, all the same. By far, an improvement over the current Gnome-Shell panel and no WM, at all of Gnome 3. It gives me worry though the Mate will drop Marco in favor of CSD. You might think "Never!" given the success that it has given to Linux Mint. But... XFCE is planning it. What's that suggest.

Last edited by Aravisian (2020-01-28 23:53:29)

Offline

#34 2020-01-28 23:58:51

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

At some point if I get chance I'll see if I can get the gtk3-mushrooms thing to work in a VM with Manjaro, it looked interesting.

Aravisian wrote:

When you spend enough time at the computer, you want to feel JOY at the eye candy when you have to stare at it instead of looking at a dull, flat, drab, uniform screen.

Yeah, I think most of us migrated to Linux because of wanting more control and flexibility in one way or another. Gnome's a bit of an aberration in that respect (at least insofar as some contributors wanting to control other people's experiences).

MATE and Cinnamon demonstrate that GTK3 can be made to work well enough for ordinary users, and Mint might be big enough to be an important player in salvaging GTK4.

With Xfce, I think it's an accumulation of factors; the window manager situation (and there seems to be a bit of internal tension there), visual preferences by a couple of devs, lack of resource, etc.

It'll be interesting to see if Mint continues with Xfce as an official spin, likewise Kali (which switched for reasons of lightness and has made a bit of a selling point out of the 'incognito' Windows 10 theme) plus other notable distro uses. It's already a bit of a strange holdover with Mint given that Cinnamon and MATE started there and that they dropped KDE, Gnome, LXDE, etc. At the time it replaced LXDE in the line-up there were effectively the two new home-grown DEs (a lightweight one and a heavier one) and two established equivalents (Xfce and KDE).

https://community.linuxmint.com/ stats note about the same number of registered Xfce as KDE users, but download stats might be a completely different story assuming they're collated.

Offline

#35 2020-02-09 12:23:01

oyvinds
Member
Registered: 2015-01-04
Posts: 11
Website

Re: Do you want to use CSD? No please!!

denyer wrote:

Chrome and other browsers do it and tend to be liked for it, but that's a rare example of utility.

One rather big difference between Chromium and GNOME "apps" is that Chromium lets you right-click on the headerbar and click "Use system title bars and borders" and then you're done with that CDN nonsense. Xfce should have a similar option. A simple [ ] Use client side decorations check-box in the window manager settings would be ideal. That way the two Xfce developers who like client side decorations can have them and all the Xfce users, many of which switched from GNOME to Xfce to avoid the silly client side decorations, are happy.

On a personal note: I use virtual desktops. I my window titlebar has that make window sticky button and I use it a lot. I also use the roll up button. Those are features that are there today. Taking them away, removing features (something GNOME is famous for), is not the right way to go.

denyer wrote:

Better known FOSS couldn't really care less about headerbars -- VLC, LibreOffice, Audacity, GIMP, FileZilla, etc. Software where most of the users aren't on Linux but on Windows, where the interfaces largely come down to traditional menus or (rarely) Microsoft's ribbons, which haven't gained wide credence either.

There is a lesson to learn from the bigger and actually successful software projects you mention: They cater to a wide audience. GNOME developers have decided that GNOME is "a platform" which is totally independent of the rest of the GNU/Linux ecosystem. GNOME "apps" are only meant to be used in the GNOME desktop environment. Useful and successful software applications can typically be used on any Linux desktop as well as Windows and macOS. None of overly simple children toy like GNOME "apps" are very successful and they never will be, their tie-in with GNOME ensures that their audience will always be limited.

Xfce should avoid making the same big mistake GNOME made and not effectively ensure that Xfce programs won't be used on any other desktop.

Reaccion Raul wrote:

I guess xfwm is kind of a mess and embracing GTK3 solves this issue. We have this visual borders since forever and It seems that this 1px zone to grab the borders is almost unfixable. I hope GTK3 solves this issue since it seems is not viable through xfwm.

The devs could ship a theme with slightly bigger borders with Xfce, problem solved. Or install one yourself. Like my yooa theme. If your distribution gave you a bad default xfce theme then that's on them and it's not a reason to switch to silly client side decorations.

oldhwman wrote:

with Gnome being such a big player, and GTK being developed, essentially, by the Gnome project and for the Gnome project, a lot of GTK applications are using CSD and header bars, for better or for worse. If you're using GTK applications, they're pretty hard to avoid by now. There aren't many developers writing applications for Linux, and many of those who do write applications that target Gnome.

No worthwhile application I'm aware of targets GNOME. Not one. Making GNOME and not Linux in general a target is dumb to begin with. KDE software leverages the KDE frameworks but they don't target KDE Plasma. There is a distinction on the KDE side between the application framework and the desktop; those are one and the same on the GNOME side. That's probably why I can't name a single GNOME "app" which isn't a complete and utter joke when it is compared to the KDE equivalent. I use quite a lot KDE programs and some non-GNOME GTK programs in Xfce. I really don't use a single GNOME "app", and as I said, I don't even know of any which are actually usable. Don't overestimate the importance of GNOME, very few would notice if that silly tablet / smart tv desktop and it's "apps" were seized to exist tomorrow.

Anyway, on a more on-topic note, here's some Xfce 4.15+ screenshots for your enjoyment: https://linuxreviews.org/Xfce_4.16

Last edited by oyvinds (2020-02-09 12:27:45)

Offline

#36 2020-02-09 17:20:34

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

Very well said, Oyvinds.
I cannot really add to the quality of your expression. But I can make a note on what you linked to. I notice that in your link it states:

The highly controversial[1][2][3] move to client side decorations do not (yet) affect software applications like the Thunar file manager, the ristretto image viewer or other built-for Xfce applications.

"Highly Controversial."
Yet, previously, Sean kept claiming that few people care about it. An opinion? Assumption? It was stated as factual; was it not checked prior to implementing this decision? That tactic is as old as human speech.
Looking further, we see that "plan is being pushed ahead anyway with all the reasons given why you better like it."
All of this is a prime example of Gnomes Years long tactics:
1.) Act like people who are not on board with your plan are "noobs" or "unenlightened" in some way. As if they don't know what is good for them.
2.) Push and bull ahead, regardless of opposition, forgetting who made your base. Gnome and now, XFCE, forget that it is not about what they want, but what the user wants. No users- No Gnome, no XFCE. They have become comfortable as an authority and we, too complacent.
3.) Removal of features in favor of Integration. This is a bad push, as Oyvinds put very well above, as it forgets the modular nature of Linux and FOSS.
Even IF CSD is made a clickable on/off choice, then it is merely a way of corralling the user base into the corner you want them to be in.
I definitely can see some merit, however, in offering a choice. The above must be noted.
4.) Leading impressions; such as implying that a change (such as switching to CSD) is "necessary" in order to achieve certain features that you may want. This is probably the most devious of all the tactics as well as untrue- but easily covered by saying, "Oh... well I didn't MEAN to give that impression..."


I believe it is time for me to look for a new Desktop Environment. I have started branching off into Marco and Cinnamon and admittedly, Linux Mint has a pretty strong reputation for assembling a solid and rugged O.S.

Last edited by Aravisian (2020-02-09 17:24:11)

Offline

#37 2020-02-10 06:12:33

oyvinds
Member
Registered: 2015-01-04
Posts: 11
Website

Re: Do you want to use CSD? No please!!

Aravisian wrote:

Yet, previously, Sean kept claiming that few people care about it. An opinion? Assumption? It was stated as factual; was it not checked prior to implementing this decision? That tactic is as old as human speech.

I suspect that seeing it as a tactic is grossly overestimating how much time and thought Sean has spent pondering up some evil scheme to push client side decorations. I suspect the truth is much simpler:

Sean has time and will to work on and improve Xfce and he does, he spends time working on it and he contributes the code and he works on what he likes to work on when he spends time on Xfce.. And he, apparently, likes client side decorations - so he writes code implementing it. I am fairly sure it is not something he does because he has this huge evil plan to hurdle Xfce users into a world filled with client side decorations.

It would be safer to assume that evil tactics are intentionally being deployed if Xfce had a multi-million dollar budget and a an authoritarian board holding secret board meetings behind closed doors on a regular basis. The simple truth is that Xfce is developed by a handful of people who work on it if/when they have time and there is no corporation or even a non-profit setup to manage any budget.

Sean does like client side decorations. Reading more than that into it seems wrong. And as I said, I think it would be optional so he can have his way while others can have theirs. Someone would, of course, have to write the code for there to be such an option. I doubt being mean to Sean will motivate him to do that.

Aravisian, you do raise many good and valid points, but I simply don't think they apply to the five people who actually work on Xfce in their spare time.

Offline

#38 2020-02-10 07:12:20

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

oyvinds wrote:
Aravisian wrote:

Yet, previously, Sean kept claiming that few people care about it. An opinion? Assumption? It was stated as factual; was it not checked prior to implementing this decision? That tactic is as old as human speech.

I suspect that seeing it as a tactic is grossly overestimating how much time and thought Sean has spent pondering up some evil scheme to push client side decorations. I suspect the truth is much simpler:

Sean has time and will to work on and improve Xfce and he does, he spends time working on it and he contributes the code and he works on what he likes to work on when he spends time on Xfce.. And he, apparently, likes client side decorations - so he writes code implementing it. I am fairly sure it is not something he does because he has this huge evil plan to hurdle Xfce users into a world filled with client side decorations.

I disagree. The fallacy of your claim is that somehow, it must take a great deal of time to jump to an assumption or use an old tactic.
This is complete nonsense. It takes almost zero time at all to spout an untested claim. People do it All The Time.
Many forums exist to painstakingly take time- to debunk the myths people propel outward in a fraction of a second.
Sean may indeed spend a great deal of time working on his projects. But your argument only demonstrates that he may have a personal bias involved- a Motive- to not bother checking whether the CSD implimentation would be well received.
What I stated was that it is an old tactic- and it is. He stated that "few people cared" and then immediately after called it "Controversial."
What I stated was that he threw that assumption out there hoping it would stick - And He did.
Nothing you just said presents any kind of challenge to what I said.

oyvinds wrote:

It would be safer to assume that evil tactics are intentionally being deployed if Xfce had a multi-million dollar budget and a an authoritarian board holding secret board meetings behind closed doors on a regular basis. The simple truth is that Xfce is developed by a handful of people who work on it if/when they have time and there is no corporation or even a non-profit setup to manage any budget.

Not at all, in fact as you like Simpler Motives: CSD lifts a lot of work off the developers allowing them to push the ball into someone else's court and remove XFWM4. Conforming to Gnomes lead has an advantage to some developers.
That is motive enough.

oyvinds wrote:

Sean does like client side decorations. Reading more than that into it seems wrong.

No, it really doesn't. I read into Exactly What He Said. His words. I quoted him.

oyvinds wrote:

I doubt being mean to Sean will motivate him to do that.

I was not being mean. I was demonstrating how Human Nature and Bias works. Whether you disagree with that is irrelevant. You see, psychology- Science- Does not care about your feelings, your bias, your hopes...Just The Facts. It is NOT mean to point out motives; it is NOT mean to question judgment; it is NOT mean to challenge a fallacy nor challenge a Misleading Statement. And it Never Should Be.
Being confrontational is not the same as being mean and calling out a fallacy is never a crime.
In fact,it is often Necessary to do so.

oyvinds wrote:

Aravisian, you do raise many good and valid points, but I simply don't think they apply to the five people who actually work on Xfce in their spare time.

I work on many projects that I put up for download- all in my spare time.
When I get feedback, I listen.
I do not tell them all the reasons why my way is better than their way.
I do not claim that the feedback is merely wrong by claiming that few people feel that way- without bothering to check how people feel about it and I certainly DO NOT claim few people feel that way and within a few sentences, call it a Controversial topic! Such a glaring contradiction demonstrates the motives of claiming that few people cared about it in the first place.


Please Do Not encourage the closure of this Valuable Feedback Thread by making claims that people are being Mean when they are not. I am in no way wrong for questioning His Own Words, nor in debating the topic by pointing out what is a fallacy.

Last edited by Aravisian (2020-02-10 07:18:33)

Offline

#39 2020-02-10 07:23:30

ozjd
Member
From: Hawkesbury NSW Australia
Registered: 2012-02-05
Posts: 560
Website

Re: Do you want to use CSD? No please!!

oyvinds wrote:

Sean does like client side decorations. Reading more than that into it seems wrong. And as I said, I think it would be optional so he can have his way while others can have theirs. Someone would, of course, have to write the code for there to be such an option. I doubt being mean to Sean will motivate him to do that.

It is actually Simon, not Sean, who is pushing this change. No one else has yet made any releases with CSD. Also there is no intention to make it optional, quoting from https://mail.xfce.org/pipermail/xfce/20 … 36807.html (worth reading the whole thread).

  including not making this optional 

Clearly not all the devs are in agreement and further discussion / dissension won't be tolerated by those behind it.

Last edited by ozjd (2020-02-10 07:24:17)

Offline

#40 2020-02-10 07:43:56

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

ozjd wrote:

It is actually Simon, not Sean, who is pushing this change. No one else has yet made any releases with CSD.

In fairness to Oyvinds, I was the one that challenged Seans words, as quoted directly from a link to discussion posted in this thread. Oyvinds did not raise Seans name needlessly, but as a direct response to my own direct words.

ozjd wrote:

Also there is no intention to make it optional, quoting from https://mail.xfce.org/pipermail/xfce/20 … 36807.html (worth reading the whole thread).

  including not making this optional 

Clearly not all the devs are in agreement and further discussion / dissension won't be tolerated by those behind it.

Wow, there is definitely tension, there.
This does not bode well...

EDIT:
I have been continuing to read. This gets interesting...
https://mail.xfce.org/pipermail/xfce/20 … 36808.html
Please read the whole thing for all context, as I will highlight some important bits here:
Above, in the link to discussion, Simon commented:

the switch to client side decorations was discussed and voted on in the
currently active core development team

This is not in the collaborative
spirit I learned when joining the project and I hope you reconsider -
especially as this is not the result of an isolated design decision taken
by an individual.

Not the collaborative spirit... Discussed and voted on...
Yet, Andrzej tells a different story:

You've mentioned discussions and voting - the only thing I've seen was a
couple of posts pretty much announcing your decision, where critical
opinions (mine included) were swept under the carpet because of "limited
manpower".

There's one motive for you, Multi-million dollar corporations unnecessary.
This is a very interesting response to a post from Simon in which Simon complained about a person not being "in the collaborative spirit". It is not in line with Simons own words but it IS in line with the general consensus of opinions in this thread and definitely my own oft-stated opinion:
That A few are Dead Set on the idea and will accept no dissension.

If this is how the leadership at XFCE development works, with the word games and hammers, I Am Not Impressed. Not One Bit.
Say one thing- Do another. Nope, not impressed.
Complain about someone not being in the collaborative spirit while hypocritically not being in the collaborative Spirit.
Nope.
Sean-Claim little dissension exists (without evidence) and then call it a highly controversial issue within the same paragraph- Nope.
Objective Eyes can see with Eyes Wide Open.

No, I am not being mean. I am upset. This is me being disturbed by the evidence that I see.

Food for thought:
IF it is true that "So Many XFCE users want CSD..." Why are they on XFCE withOUT CSD instead of being on Gnome which is chock full of CSD?
"Most of the people in the Vatican want to switch to Islam." "Then why are they in the Vatican instead of in the Mosque?"
Just sayin'...

Last edited by Aravisian (2020-02-10 09:33:29)

Offline

#41 2020-02-10 09:50:35

oyvinds
Member
Registered: 2015-01-04
Posts: 11
Website

Re: Do you want to use CSD? No please!!

ozjd wrote:

It is actually Simon, not Sean, who is pushing this change. No one else has yet made any releases with CSD.

So sorry for mixing Simon Steinbeiß up with Sean M. Davis (bluesabre). The post I quoted said Sean and I didn't catch that he had mixed them up so I did too. Very sorry.

ozjd wrote:

Clearly not all the devs are in agreement and further discussion / dissension won't be tolerated by those behind it.

I kind of doubt I'd be kicked off #xfce-dev if I bring it up. Perhaps I will, politely, do that just to see what happens (it's an efficient way to measure how authoritarian a free software projects leadership is).

Based Andrzej brings up an interesting point in https://mail.xfce.org/pipermail/xfce/20 … 36808.html

Based Andrzej wrote:

"You've mentioned discussions and voting - the only thing I've seen was a couple of posts pretty much announcing your decision, where critical  opinions (mine included) were swept under the carpet because of "limited  manpower"."

There is also an interesting and perhaps more important point in Simon Steinbeiß's message at https://mail.xfce.org/pipermail/xfce/20 … 36807.html

Simon Steinbeiß wrote:

"We didn't take this decision lightly and there was a lot of discussion, but in the end we agreed on taking this step - including not making this optional (in parts because of the very limited manpower the project has at hand). I stand by this common decision."

My humble opinion is that if Simon Steinbeiß wants to implement support for CSD's and doesn't want to or have time to write code which makes CSD optional then that's fine. They do have limited manpower and he is, and should be, free to spend his valuable time on whatever he feels like coding.

Now, the obvious question is: What happens if someone does step up with a patch which makes those CSD's optional? If such a patch is rejected GNOME-style with a "No, we don't want our users to have that feature" then it's time to abandon Xfce and pick some alternative like KDE, LXQt or IceWM. That is, for now, a rather hypothetical question since nobody has written or submitted such a patch. I don't think we are entitled to whine too much until or unless such a patch is proposed. That's when we get to see if Aravisian's suspicions that there a The Evil XFCE CDN Conspiracy are true or if this is simply a matter of very limited developer time/resources.

Aravisian wrote:

IF it is true that "So Many XFCE users want CSD..." Why are they on XFCE withOUT CSD instead of being on Gnome which is chock full of CSD?

That is a very good point: A lot of Xfce users are using it because they don't want GNOME and GNOME nonsense like client side decorations.

One slightly related point: GNOME will always be better than XFCE at being a silly useless CPU-cycle-wasting tablet/smart-TV desktop joke of a desktop. XFCE can try to copy GNOME and add client side decorations, remove the system tray and other features and waste everyone's time - but XFCE will never be GNOME or be better than GNOME at being GNOME. XFCE can, on the other hand, be very good at being a sane light-weight traditional desktop environment. It's already good at that and that's also the only thing it's good at. I would prefer if they don't take that away.

Last edited by oyvinds (2020-02-10 09:52:09)

Offline

#42 2020-02-10 10:03:43

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

oyvinds wrote:
ozjd wrote:

It is actually Simon, not Sean, who is pushing this change. No one else has yet made any releases with CSD.

So sorry for mixing Simon Steinbeiß up with Sean M. Davis (bluesabre). The post I quoted said Sean and I didn't catch that he had mixed them up so I did too. Very sorry.

I did not mix them up.
I quoted Seans comments on the matter Directly.
They were not mixed up by me or by you. I addressed Seans comments. Just to be clear.

oyvinds wrote:

My humble opinion is that if Simon Steinbeiß wants to implement support for CSD's and doesn't want to or have time to write code which makes CSD optional then that's fine. They do have limited manpower and he is, and should be, free to spend his valuable time on whatever he feels like coding.

True.
Yet it is also true that the code he writes for a purpose- that has a user base, is based on that code HAVING a user base.
I agree, if he doesn't like it, He can Quit.
No one is forcing him to work on XFCE. However... the Users are the one that decides whether that code is used o their computers. So in the end...

oyvinds wrote:

I don't think we are entitled to whine too much until or unless such a patch is proposed. That's when we get to see if Aravisian's suspicions that there a The Evil XFCE CDN Conspiracy are true or if this is simply a matter of very limited developer time/resources.

Whine. Evil. Cut the Word Spins.
It is Not Whining to speak up. That is like calling anyone that diesagrees with YOUR point of view a Noob, or an Idiot or any other Ad Hominem insult you care to throw against your opponents.
I never called them Evil.
I called them deceptive according to their Bias and Human Condition.
I never once described any conspiracy. Keep your directions toward me Accurate. I described them being Stubborn and bulling forward despite opposition. That is not what defines a conspiracy.

oyvinds wrote:

Aravisian wrote:

    IF it is true that "So Many XFCE users want CSD..." Why are they on XFCE withOUT CSD instead of being on Gnome which is chock full of CSD?

That is a very good point: A lot of Xfce users are using it because they don't want GNOME and GNOME nonsense like client side decorations.

Yes, it has merit as a proposed assumption. It's a bit of a stand off sicne we do not have a valid Polling to measure. However, that valid point above must carry more weight than the assumption that many to most users want CSD by logic, alone.

oyvinds wrote:

One slightly related point: GNOME will always be better than XFCE at being a silly useless CPU-cycle-wasting tablet/smart-TV desktop joke of a desktop. XFCE can try to copy GNOME and add client side decorations, remove the system tray and other features and waste everyone's time - but XFCE will never be GNOME or be better than GNOME at being GNOME. XFCE can, on the other hand, be very good at being a sane light-weight traditional desktop environment. It's already good at that and that's also the only thing it's good at. I would prefer if they don't take that away.

How close can you get before it becomes "close enough?"

Honestly... after reading the above Links... I wouldn't use the word 'sane'. Repeating mistakes made by others expecting different results does not really fall into the 'sane' category.

Last edited by Aravisian (2020-02-10 10:06:40)

Offline

#43 2020-02-10 18:55:09

mauser
Member
Registered: 2019-01-23
Posts: 37

Re: Do you want to use CSD? No please!!

That's what I will do if the Brisk menu in Mate can be resized if Xfce goes with CSD or I will go with KDE. --snip--

sixsixfive wrote:

Just let the developers do what they want - If you dont want CSDs use

* MATE
* UKUI
* LXQt
* KDE

you can also use plenty of window managers with Qt or non-CSD GTK apps

----------------------------------------------------------------------------------------------

I personally switched to MATE a year ago when all this was beginning with catfish or the taskmanager.

And i must really say MATE is a lot better than I remember GNOME2(which I never liked - damn i miss KDE3), I still use thunar or pcmanfm for file management since caja is terrible slow at image previews.

But I really like that MATE disabled the overlay scrolling and the dialog headerbars by deault. Not to mention that it has a polkit client, archive manager and atril. All MATE apps have a traditional style: menubar, toolbar and normal menus with icons (not those ugly osx-like popovers).

PS: I have used gtk3-nocsd in the past but this is a dead end since GTK4 won't support modules anymore.

Last edited by ToZ (2020-02-14 06:22:29)


I am command line illiterate. I use MX-21.3 Linux.

Offline

#44 2020-02-10 20:04:01

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

oyvinds wrote:

One rather big difference between Chromium and GNOME "apps" is that Chromium lets you right-click on the headerbar and click "Use system title bars and borders" and then you're done with that

True.

https://mail.xfce.org/pipermail/xfce/20 … 36808.html

Do we have any idea who's classed as the "currently active core development team"? Andrzej gets shut down in that discussion without any elaboration on who or how many voted.

In terms of public posters to the Xfce blog, core is Sean, Simon and occasionally André. Git suggests a handful of other names. The Wiki's mainly Simon. Bountysource activity seems to basically be Sean. Twitter Sean/Simon. Bugzilla has more activity on particular components, but mostly from reporters.

If it's already gotten to the point of other people's code being edited and that being justified by referring to a secret ballot which excluded other developers, things aren't in a healthy state, it goes without saying.

Conspiracy is too grand a word, but it does look like opinions were only solicited to reinforce a decision, and that there's no interest in transparency or having governance structures or user feedback mechanisms that involve transparency and might demonstrate otherwise.

oyvinds wrote:

What happens if someone does step up with a patch which makes those CSD's optional?

With the type of changes it's more a case of fork from 4.14 and go from there, retaining existing code and adding in the new stuff if wanted. Define a global setting and use it for if X then Y else Z constructs, effectively maintaining two interfaces. Essentially what's happened with Catfish, except that's per-app rather than global.

As before, interested how this is going to play out elsewhere (Debian being conservative and emphasising standards, Manjaro having had some pushback against CSD, Xfce components being used in other environments, etc). Breaking with standards is a big decision for a DE that prioritises standards and is often selected on that basis.

Offline

#45 2020-02-10 21:19:15

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

denyer wrote:

Do we have any idea who's classed as the "currently active core development team"? Andrzej gets shut down in that discussion without any elaboration on who or how many voted.
If it's already gotten to the point of other people's code being edited and that being justified by referring to a secret ballot which excluded other developers, things aren't in a healthy state, it goes without saying.

Couldn't agree more. Even if the developers were to agree to scrap their plans for now, it would leave a sour taste. And, it won't be long before they swing back to the idea.

denyer wrote:

Conspiracy is too grand a word, but it does look like opinions were only solicited to reinforce a decision, and that there's no interest in transparency or having governance structures or user feedback mechanisms that involve transparency and might demonstrate otherwise.

Excellent rebuttal to some Quixotic tilting at windmills. You handled that far better than I did.

denyer wrote:

With the type of changes it's more a case of fork from 4.14 and go from there, retaining existing code and adding in the new stuff if wanted. Define a global setting and use it for if X then Y else Z constructs, effectively maintaining two interfaces. Essentially what's happened with Catfish, except that's per-app rather than global.

As before, interested how this is going to play out elsewhere (Debian being conservative and emphasising standards, Manjaro having had some pushback against CSD, Xfce components being used in other environments, etc). Breaking with standards is a big decision for a DE that prioritises standards and is often selected on that basis.

The problem with an Optional CSD is that it is counterproductive to what the developers want to do. It is being shot down for a reason: If they implement that, they may as well scrap the whole idea.
See, part of "Their" decision... is that CSD will entirely replace the WM, just as in the case of Gnome. Their plan is to get rid of the XFWM, dropping that project from their dockets.
If they make it optional to have CSD, then it is pointless to have CSD because no one REALLY wants it and it is already an option- just use Gnome. The only reason it becomes viable for the developers is if it allows them to Drop the Window manager. No Window Manager, no option to turn CSD off.
This is how Gnome is as well and why gtk3-nocsd only works on Non-Gnome desktops that have a window manager, like XFCE is currently.
Gtk-nocsd will NOT work on XFCE 4.16, for example, because there is no window manager to support it and therefor, no other option to disable CSD.

Last edited by Aravisian (2020-02-10 21:22:42)

Offline

#46 2020-02-10 22:30:28

jakfish
Member
Registered: 2019-12-23
Posts: 55

Re: Do you want to use CSD? No please!!

>> Gtk-nocsd will NOT work on XFCE 4.16, for example, because there is no window manager to support it and therefor, no other option to disable CSD. <<

Ah. I was just about to post and say, well, okay, what about gtk-nocsd? I suppose a user could freeze any xfce updates past 4.14, but the trouble with that, obviously, is a new install from a current iso will give you 4.16 for starters, and I don't know: 1) if you can downgrade to 4.14  2) how long a user can limp along with 4.14 and dist-upgrade before something bad happens.

Jake

Last edited by jakfish (2020-02-10 22:33:08)

Offline

#47 2020-02-10 22:54:13

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

jakfish wrote:

>> Gtk-nocsd will NOT work on XFCE 4.16, for example, because there is no window manager to support it and therefor, no other option to disable CSD. <<

Ah. I was just about to post and say, well, okay, what about gtk-nocsd? I suppose a user could freeze any xfce updates past 4.14, but the trouble with that, obviously, is a new install from a current iso will give you 4.16 for starters, and I don't know: 1) if you can downgrade to 4.14  2) how long a user can limp along with 4.14 and dist-upgrade before something bad happens.

Jake

This is a good point. yes, you could use "apt-mark hold" for at least a few years. But past that, you may start running into trouble.
However, if the current version of XFCE is fissioned off into a new desktop- then it could continue on.  While I am not necessarily encouraging a mutiny on the Bounty... OK... Wait, yeah, actually I would be totally fine with it.
Yes, you can 'downgrade' (I'd call it an upgrade!) to 4.14 from 4.16 as long as you have the packages available to install--- I downgraded Zorin 15.1 Lite from 4.14 to 4.12 and it was not difficult to do. And it came with the bonus of helping me to learn much more about how XFCE works as a package. Might come in handy, knowing that... some day.
When XFCE 4.14 as a version is no longer maintained by anyone is when it starts to falter. With many Big Name desktops with a user base of their own using XFCE, now, they may not want to deal with the flack of user feedback and take up the slack, much as how Mate fissioned off as a continuation of Gnome2.
And it is still around and Thriving.
I have even wondered if that is what the XFCE developers are hoping for- allowing them to handle a smaller project and allowing someone else to take up the mantle of XFCE as it is now.
Since switching to CSD means dropping the XFWM, then having an option to turn off or on the CSD means keeping the XFWM so instead of lightening the load, adding CSD would only be adding to the system instead of reducing it.

Last edited by Aravisian (2020-02-10 23:22:18)

Offline

#48 2020-02-11 03:26:48

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

I'd be surprised if much was actively hacked out of Xfwm4 in the immediate future.

Further down the line? A more traditional-looking Elementary style environment seems to be the intended destination.

Aravisian wrote:

I have even wondered if that is what the XFCE developers are hoping for- allowing them to handle a smaller project and allowing someone else to take up the mantle of XFCE as it is now

Certainly wouldn't be a huge surprise if Mint holds at 4.14 for a relatively long time, and if a fork was made Wayland isn't a big priority for them. There's precedent with things like Nemo and Xapps. But they may take the view that as the focus is on their own apps, and they do already bundle some CSD apps, that it isn't a major issue.

Would tend to doubt it's something the Xfce devs steering these changes want.

Interesting discussion elsewhere: https://forum.mxlinux.org/viewtopic.php?f=93&t=56087

Offline

#49 2020-02-11 03:40:28

Aravisian
Member
Registered: 2019-08-17
Posts: 410

Re: Do you want to use CSD? No please!!

denyer wrote:

I'd be surprised if much was actively hacked out of Xfwm4 in the immediate future.

Further down the line? A more traditional-looking Elementary style environment seems to be the intended destination.

That is not what the take-away from the blog announcement is. It quite clearly stated that XFWM was slated for removal.

denyer wrote:

Would tend to doubt it's something the Xfce devs steering these changes want.

YOu could be right- I just wondered if it was part of the package; by dumping XFWM and picking up CSD< they could reduce their workload. As they stated they have limited manpower.
It seemed possible that they were hoping someone else would pickup and carry the slack. However, while this could allow XFCE to move in a whole new direction (Gnome Clone) while a fork that embodies the current and previous XFCE4 could be competition that new direction, leaving them competing with what used to be their project. You just may have a point that they may not want that.

Offline

#50 2020-02-11 09:06:46

denyer
Member
From: UK
Registered: 2018-10-20
Posts: 63
Website

Re: Do you want to use CSD? No please!!

Which bit of which announcement are you referring to? Xfwm4 handles non-Xfce applications too and will be needed in some form for years. (X support being dropped in desktop environments is a long way away for most of them).

It may not have the same feature set but it'll be around.

Last edited by denyer (2020-02-11 09:08:13)

Offline

Board footer

Powered by FluxBB